Confusing crowds with communities

(Life of Brian on crowds)

This morning I attended the Social Media Week session “Open UN: The evolution of the crowd” (see programme and video of the event here on the UN Pulse website).

I was particularly interested in this panel since I’ve had a long standing interest in “the wisdom of crowds” and how this might be applied in development which predates my current work on knowledge management and communities of practice.

Listening to the panel discussion though I quickly realized that while the panel was notionally about crowds, in fact what most of the panelists were describing were the creation of communities, not crowds at all.  Mark Belinski of Digital Democracy even went as far as to say “Communities share info, crowds make noise”.

If we are to think about “listening to the crowd” and “crowdsourcing ” then it’s important to have a clearer understanding of the differences between crowds and communities where each occur and how we can “use” them for knowledge sharing for development.

A community is a group of people who come together to act around a common purpose or “intent”. It is also characterized by interactions and relationships between its members who are in some way dependent on one another. A community may have formalized organization and rules, or it may only have informal norms, it may be intentionally designed or it may “emerge” spontaneously. It may have clear leadership or it may have highly decentralized leadership. As I mentioned in a previous post – even if a community emerges spontaneously, over time it will develop its own rules, norms and heirarchies and methods of co-ordination. It’s not a flat leaderless structure where all members have equal influence.

A crowd by contrast is made up of individuals each pursuing their own interests. While information might emerge from a crowd it is the result of individual actions (and often each member of the crowd watching and reacting to what the others are doing) but it doesn’t have a conscious “purpose” or organization. An example might be the stock market where stock prices are set by the “crowd” of investors each making their own decisions pursuing their own goals, but from which information and order (in this case a stock price) emerges.

Often communities can emerge from crowds but they are not the same thing, and that matters. Here are a few of the implications of the differences.

1. People choose to be in communities. Communities  can represent “local ownership and participation” in the development process, and an opportunity for aid beneficiaries to have a voice. It also provides development workers with clear interlocateurs to interact with. It’s therefore clearly desirable to help foster communities and provide the means for them to self-organize and communicate and take action. BUT we should also recognize that communities establish rules and hierarchies even if these evolve over time and are not formal or explicit. This means that not all voices in a community are equal. Some members have more influence than others, and some make a greater contribution to the cause than others. Not all members of a group of beneficiaries will necessarily be willing or able to join a community or play an active part in it, and not all voices or interests are treated equally. Those with more personal influence and those who are able to put in most effort will also have more influence over the community and most likely benefit most from it.

2. Since  crowds are more passive or “purposeless” then you can’t interact with them as if they have a common voice. But a crowd can include members of beneficiary groups that don’t actively participate in any communities. Here the means to collect information from the crowd is also more passive, and it’s here where a broader sense of “information exhaust” can come into play. To understand a crowd you need to try to observe what a crowd does, and aggregate it into meaningful data. This can be by collecting and analyzing lots of micro level data on behaviour such as visits to health centres, mobile phone usage, school attendance etc. and then using analytical techniques to make sense of it. It can include collecti0n systems that require participation such as voting or opinion polls, or SMS alert systems such as Ushahidi – but in this case we need to be careful to ensure the barriers to participation are not so high as to bias the information collected. In a crowd situation generally one voice is not more important than another, or if it is this is not due to the communication and leadership skills of the contributor. Observing crowds can also be a better way to see divergent or minority views or situations within a population whereas communities often promote convergence around a single viewpoint.

A key difference here is that we collect data from crowds and interpret it ourselves often using analytical techniques to make sense of it, whereas in a community, the community itself determines the “meaning” of the information collected. Crowd data could be thought of as more “objective” since it is is based on observation, whereas community data is more subjective as it is based on the communities interpretation of its own situation, but it also gives a more contextualized understanding of the local situation.

But both of these approaches can be very valuable for development giving complementary views of the same situation by bringing together observation and participation. It’s not a question of communities versus crowds but rather making best use of each to better understand a situation, engage with beneficiaries and work together to take appropriate action.